Project Summary Over the last few years, it has become evident that a significant proportion of human genes are associated with their respective natural antisense transcript, which overlaps the protein coding RNA transcribed in the sense direction. A body of evidence has emerged demonstrating that these long antisense non-coding transcripts actively participate in the regulation of protein-coding sense RNA transcription and RNA processing at various levels. The widespread occurrence of antisense transcription documents the prevalence of gene regulation by natural antisense transcripts. Antisense non-coding RNAs have been implicated in the silencing of tumor suppressor genes. The recent studies by our group (2) as well as others (3) suggest that the epigenetic silencing of tumor suppressor genes is the result of an imbalance between the levels of bidirectional, i.e. sense and antisense, transcription. Here, we propose to study the roles of long non-coding antisense RNAs in gene silencing and in gene activation. We will identify and characterize long non-coding transcripts for representative silenced tumor suppressor genes. We will then re-activate the transcription of silenced tumor suppressors by interfering with the activity of antisense transcripts and determine the molecular mechanism of this re- activation. We will also determine the effects of tumor suppressor re-activation on the cellular phenotype. This work will increase our understanding of tumor suppressor silencing and by reverting the silenced state will provide new therapeutic and diagnostic approaches.